Organizational Choice:

Product vs. Function

Arthur H. Walker and Jay W. Lorsch

No. 68613
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
 NOVEMBER–DECEMBER 1968
Organizational Choice:
Product vs. Function
Arthur H. Walker and Jay W. Lorsch
Corporations, especially manufacturers, long have
wrestled with the problem of how to structure their orga-
nizations to enable employees, particularly the specialists,
to do their jobs with maximum efficiency and productiv-
ity. One perplexing issue has been whether to organize
around functions or products. Here, two behavioral scien-
tists look at the question in light of recent studies. They
focus on two plants (of two of the largest consumer prod-
ucts companies), one organized by product, the other by
function.
O f all the issues facing a manager as he thinks
about the form of his organization, one of the
thorniest is the question of whether to group activi-
ties primarily by product or by function. Should all
specialists in a given function be grouped under a
common boss, regardless of differences in products
they are involved in, or should the various functional
specialists working on a single product be grouped
together under the same superior?
Mr. Walker is Associate Professor of Management at
Northeastern University, where he teaches courses in human
relations and organizational behavior to undergraduates and
graduate students. Before turning to academic life, he spent 15
years in industry. Mr. Lorsch, coauthor of two previous HBR arti-
cles, is Associate Professor of Organizational Behavior at the
Harvard Business School, and his principal interest is the struc-
ture of organizations in relation to their environment.
Copyright © 1968 by the President and Fellows of Harvard College. All rights reserved.
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
In talks with managers we have repeatedly heard
them anguishing over this choice. For example,
recently a divisional vice president of a major U.S. cor-
poration was contemplating a major organizational
change. After long study, he made this revealing
observation to his subordinate managers:
“We still don’t know which choice will be the best
one. Should the research, engineering, marketing, and
production people be grouped separately in depart-
ments for each function? Or would it be better to have
them grouped together in product departments, each
department dealing with a particular product group?
“We were organized by product up until a few years
ago. Then we consolidated our organization into spe-
cialized functional departments, each dealing with all
of our products. Now I’m wondering if we wouldn’t be
better off to divide our operations again into product
units. Either way I can see advantages and disadvan-
tages, trade-offs. What criteria should I use? How can
we predict what the outcomes will be if we change?”
Companies that have made a choice often feel con-
fident that they have resolved this dilemma. Consider
the case of a large advertising agency that consolidated
its copy, art, and television personnel into a “total cre-
ative department.” Previously they had reported to
group heads in their areas of specialization. In a memo
to employees the company explained the move:
 “Formation of the ‘total creative’ department com-
pletely tears down the walls between art, copy, and
television people. Behind this move is the realization
that for best results all creative people, regardless of
their particular specialty, must work together under
the most intimate relationship as total advertising
people, trying to solve creative problems together
from start to finish.
“The new department will be broken into five
groups reporting to the senior vice president and cre-
ative director, each under the direction of an associ-
ate creative director. Each group will be responsible
for art, television, and copy in their accounts.”
But our experience is that such reorganizations
often are only temporary. The issues involved are so
complex that many managements oscillate between
these two choices or try to effect some compromise
between them.
In this article we shall explore—from the viewpoint
of the behavioral scientist—some of the criteria that
have been used in the past to make these choices, and
present ideas from recent studies that suggest more
relevant criteria for making the decision. We hope to
provide a way of thinking about these problems that
will lead to the most sensible decisions for the accom-
plishment of organizational goals.
The dilemma of product versus function is by no
means new; managers have been facing the same
basic question for decades. As large corporations like
Du Pont and General Motors grew, they found it nec-
essary to divide their activities among product divi-
sions.1 Following World War II, as companies ex-
panded their sales of existing products and added new
products and businesses, many of them implemented
a transition from functional organizations handling a
number of different products to independently man-
aged product divisions. These changes raised prob-
lems concerning divisionalization, decentralization,
corporate staff activities, and the like.
As the product divisions grew and prospered,
many companies extended the idea of product orga-
nization further down in their organizations under
such labels as “the unit management concept.”
Today most of the attention is still being directed to
these changes and innovations within product or
market areas below the divisional level.
We are focusing therefore on these organizational
issues at the middle and lower echelons of manage-
ment, particularly on the crucial questions being
faced by managers today within product divisions.
The reader should note, however, that a discussion
of these issues is immensely complicated by the fact
that a choice at one level of the corporate structure
1. For a historical study of the organizational structure of U.S. corpora-
tions, see Alfred D. Chandler, Jr., Strategy and Structure (Cambridge, The
M.I.T. Press, 1962).
130
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
affects the choices and criteria for choice at other
levels. Nonetheless, the ideas we suggest in this
article are directly relevant to organizational choice
at any level.
Elements to Consider
To understand more fully the factors that make
these issues so difficult, it is useful to review the
criteria often relied on in making this decision.
Typically, managers have used technical and eco-
nomic criteria. They ask themselves, for instance,
“Which choice will minimize payroll costs?” Or,
“Which will best utilize equipment and special-
ists?” This approach not only makes real sense in
the traditional logic of management, but it has
strong support from the classical school of organi-
zation theorists. Luther Gulick, for example, used it
in arguing for organization by function:
“It guarantees the maximum utilization of up-to-
date technical skill and . . . makes it possible in
each case to make use of the most effective divi-
sions of work and specialization. . . . [It] makes
possible also the economies of the maximum use
of labor-saving machinery and mass produc-
tion. . . . [It] encourages coordination in all of the
technical and skilled work of the enterprise. . . .
[It] furnishes an excellent approach to the develop-
ment of central coordination and control.”2
In pointing to the advantages of the product basis
of organization, two other classical theorists used
the same approach:
“Product or product line is an important basis for
departmentalizing, because it permits the maximum
use of personal skills and specialized knowledge,
facilitates the employment of specialized capital and
makes easier a certain type of coordination.”3
In sum, these writers on organization suggested
that the manager should make the choice based on
three criteria:
1. Which approach permits the maximum use of
special technical knowledge?
2. Which provides the most efficient utilization
of machinery and equipment?
3. Which provides the best hope of obtaining the
required control and coordination?
There is nothing fundamentally wrong with these
criteria as far as they go, and, of course, managers
2. Luther Gulick, “Notes on the Theory of Organization,” in Papers on
the Science of Administration, edited by Luther Gulick and Lyndall F.
Urwick (New York, New York Institute of Public Administration, 1937),
pp. 23–24.
3. Harold D. Koontz and C.J. O’Donnell, Principles of Management (New
York, McGraw-Hill Book Company, Inc., 2nd edition, 1959), p. 111.
HARVARD BUSINESS REVIEW November–December 1968
 have been using them. But they fail to recognize the
complex set of trade-offs involved in these decisions.
As a consequence, managers make changes that pro-
duce unanticipated results and may even reduce the
effectiveness of their organization. For example:
M A major manufacturer of corrugated containers a
few years ago shifted from a product basis to a func-
tional basis. The rationale for the decision was that it
would lead to improved control of production costs
and efficiencies in production and marketing. While
the organization did accomplish these aims, it found
itself less able to obtain coordination among its local
sales and production units. The functional special-
ists now reported to the top officers in charge of pro-
duction and sales, and there was no mechanism for
one person to coordinate their work below the level
of division management. As a result, the company
encountered numerous problems and unresolved
conflicts among functions and later returned to the
product form.
This example pinpoints the major trade-off that
the traditional criteria omit. Developing highly spe-
cialized functional units makes it difficult to
achieve coordination or integration among these
units. On the other hand, having product units as
the basis for organization promotes collaboration
between specialists, but the functional specialists
feel less identification with functional goals.
Behaviorists’ findings
We now turn to some new behavioral science
approaches to designing organization structure.
Recent studies4 have highlighted three other impor-
tant factors about specialization and coordination:
M As we have suggested, the classical theorists saw
specialization in terms of grouping similar activities,
skills, or even equipment. They did not look at its psy-
chological and social consequences. Recently, behav-
ioral scientists (including the authors) have found that
there is an important relationship between a unit’s or
individual’s assigned activities and the unit members’
patterns of thought and behavior. Functional special-
ists tend to develop patterns of behavior and thought
that are in tune with the demands of their jobs and
their prior training, and as a result these specialists
(e.g., industrial engineers and production supervisors)
have different ideas and orientation about what is
important in getting the job done. This is called dif-
4. See Paul R. Lawrence and Jay W. Lorsch, Organization and Environment
(Boston, Division of Research, Harvard Business School, 1967); and Eric J.
Miller and A.K. Rice, Systems of Organization (London, Tavistock
Publications, 1967).
HARVARD BUSINESS REVIEW November–December 1968
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
ferentiation, which simply means the differences in
behavior and thought patterns that develop among
different specialists in relation to their respective
tasks. Differentiation is necessary for functional spe-
cialists to perform their jobs effectively.
M Differentiation is closely related to achievement
of coordination, or what behavioral scientists call
integration. This means collaboration between spe-
cialized units or individuals. Recent studies have
demonstrated that there is an inverse relationship
between differentiation and integration: the more
two functional specialists (or their units) differ in
their patterns of behavior and thought, the more dif-
ficult it is to bring about integration between them.
Nevertheless, this research has indicated, achieve-
ment of both differentiation and integration is essen-
tial if organizations are to perform effectively.
M While achievement of both differentiation and
integration is possible, it can occur only when well-
developed means of communication among special-
ists exist in the organization and when the specialists
are effective in resolving the inevitable cross-func-
tional conflicts.
These recent studies, then, point to certain related
questions that managers must consider when they
choose between a product or functional basis of
organization:
1. How will the choice affect differentiation among
specialists? Will it allow the necessary differences in
viewpoint to develop so that specialized tasks can be
performed effectively?
2. How does the decision affect the prospects of
accomplishing integration? Will it lead, for instance,
to greater differentiation, which will increase the
problems of achieving integration?
3. How will the decision affect the ability of orga-
nization members to communicate with each other,
resolve conflicts, and reach the necessary joint
decisions?
There appears to be a connection between the
appropriate extent of differentiation and integration
and the organization’s effectiveness in accomplishing
its economic goals. What the appropriate pattern is
depends on the nature of external factors—markets,
technology, and so on—facing the organization, as
well as the goals themselves. The question of how the
organizational pattern will affect individual members
is equally complex. Management must consider how
much stress will be associated with a certain pattern
and whether such stress should be a serious concern.
To explore in more detail the significance of mod-
ern approaches to organizational structuring, we
shall describe one recent study conducted in two
131
 manufacturing plants—one organized by product,
the other on a functional basis.5
Plant F & Plant P
The two plants where this study was conducted
were selected because they were closely matched in
several ways. They were making the same product;
their markets, technology, and even raw materials
were identical. The parent companies were also sim-
ilar: both were large, national corporations that
developed, manufactured, and marketed many con-
sumer products. In each case divisional and corpo-
rate headquarters were located more than 100 miles
from the facilities studied. The plants were sepa-
rated from other structures at the same site, where
other company products were made.
Both plants had very similar management styles.
They stressed their desire to foster employees’ ini-
tiative and autonomy and placed great reliance on
selection of well-qualified department heads. They
also identified explicitly the same two objectives.
The first was to formulate, package, and ship the
products in minimum time at specified levels of
quality and at minimum cost—that is, within exist-
ing capabilities. The second was to improve the
capabilities of the plant.
In each plant there were identical functional spe-
cialists involved with the manufacturing units and
packing unit, as well as quality control, planning and
scheduling, warehousing, industrial engineering, and
plant engineering. In Plant F (with the functional
basis of organization), only the manufacturing depart-
ments and the planning and scheduling function
reported to the plant manager responsible for the
product (see Exhibit I). All other functional special-
ists reported to the staff of the divisional manufactur-
ing manager, who was also responsible for plants
manufacturing other products. At Plant P (with the
product basis of organization), all functional special-
ists with the exception of plant engineering reported
to the plant manager (see Exhibit II).
State of differentiation
In studying differentiation, it is useful to focus on
the functional specialists’ differences in outlook in
terms of:
V Orientation toward goals.
V Orientation toward time.
V Perception of the formality of organization.
5. Arthur H. Walker, Behavioral Consequences of Contrasting Patterns of
Organization (Boston, Harvard Business School, unpublished doctoral dis-
sertation, 1967).
132
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
Goal orientation: The bases of organization in the
two plants had a marked effect on the specialists’ dif-
ferentiated goal orientations. In Plant F they focused
sharply on their specialized goals and objectives. For
example, quality control specialists were concerned
almost exclusively with meeting quality standards,
industrial engineers with methods improvements
and cost reduction, and scheduling specialists with
how to meet schedule requirements. An industrial
engineer in Plant F indicated this intensive interest
in his own activity:
“We have 150 projects worth close to a million
dollars in annual savings. I guess I’ve completed
some that save as much as $90,000 a year. Right now
I’m working on cutting departmental costs. You
need a hard shell in this work. No one likes to have
his costs cut, but that is my job.”
That these intense concerns with specialized
objectives were expected is illustrated by the apolo-
getic tone of a comment on production goals by an
engineering supervisor at Plant F:
“At times we become too much involved in pro-
duction. It causes a change in heart. We are inter-
ested in production, but not at the expense of our
own standards of performance. If we get too much
involved, then we may become compromised.”
A final illustration is when production employees
stood watching while members of the maintenance
department worked to start a new production line,
and a production supervisor remarked:
“I hope that they get that line going soon. Right
now, however, my hands are tied. Maintenance has
the job. I can only wait. My people have to wait,
too.”
This intense concern with one set of goals is anal-
ogous to a rifle shot; in a manner of speaking, each
specialist took aim at one set of goals and fired at it.
Moreover, the specialists identified closely with
their counterparts in other plants and at divisional
headquarters. As one engineer put it:
“We carry the ball for them (the central office). We
carry a project through and get it working right.”
At Plant P the functional specialists’ goals were
more diffuse—like buckshot. Each specialist was
concerned not only with his own goals, but also
with the operation of the entire plant. For example,
in contrast to the Plant F production supervisor’s
attitude about maintenance, a Plant P maintenance
manager said, under similar circumstances:
“We’re all interested in the same thing. If I can
help, I’m willing. If I have a mechanical problem,
there is no member of the operating department who
wouldn’t go out of his way to solve it.”
Additional evidence of this more diffuse orientation
toward goals is provided by comments such as these
which came from Plant P engineers and managers:
HARVARD BUSINESS REVIEW November–December 1968
 EXHIBIT I. Organizational Chart at Plant F

Manufacturing
general
manager

Plant Industrial Warehouse/ Industrial Quality

Other Traffic Plant
accounting relations supply engineering control
staff manager engineer
manager manager manager manager manager

Plant F
plant
manager

Scheduling
Special Processing Processing Processing and Quality
Sanitation Maintenance
assignment unit 1 unit 2 unit 3 materials control
planning

“We are here for a reason—to run this place the best
way we know how. There is no reluctance to be open
and frank despite various backgrounds and ages.”
“The changeovers tell the story. Everyone shows
willingness to dig in. The whole plant turns out to
do cleaning up.”
Because the functional specialists at Plant F focused
on their individual goals, they had relatively wide dif-
ferences in goals and objectives. Plant P’s structure,
on the other hand, seemed to make functional spe-
cialists more aware of common product goals and
reduced differences in goal orientation. Yet, as we
shall see, this lesser differentiation did not hamper
their performance.
Time orientation: The two organizational bases
had the opposite effect, however, on the time orien-
tation of functional managers. At Plant F, the spe-
cialists shared a concern with short-term issues
(mostly daily problems). The time orientation of
specialists at Plant P was more differentiated. For
example, its production managers concentrated on
routine matters, while planning and industrial engi-
neering focused on issues that needed solution
within a week, and quality control specialists wor-
ried about even longer-term problems.
The reason is not difficult to find. Since Plant P’s
organization led its managers to identify with prod-
uct goals, those who could contribute to the solu-
tion of longer-term problems became involved in
these activities. In Plant F, where each unit focused
on its own goals, there was more of a tendency to
worry about getting daily progress. On the average,
employees of Plant P reported devoting 30% of their
time to daily problems, while at Plant F this figure
was 49%. We shall have more to say shortly about
how these factors influenced the results achieved in
the two plants.
Organizational formality: In the study, the for-
mality of organizational structure in each functional
activity was measured by three criteria:
1. Clarity of definition of job responsibilities.
2. Clarity of dividing lines between jobs.
3. Importance of rules and procedures.
It was found that at Plant F there were fewer dif-
ferences among functional activities in the formal-
ity of organization structure than at Plant P. Plant F
employees reported that a uniform degree of struc-
ture existed across functional specialties; job respon-

HARVARD BUSINESS REVIEW November–December 1968 133

This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
 EXHIBIT II. Organizational Chart at Plant P

Manufacturing
general
manager

Plant Plant
Special Personnel Plant
industrial chemical
assignment manager engineer
engineer engineer

Project
engineer

Plant P
plant
manager

Processing Packaging Warehouse Industrial Quality

Maintenance
unit unit and supply engineer control

sibilities were well defined, and the distinctions
between jobs were clear. Similarly, rules and proce-
dures were extensively relied on. At Plant P, on the
other hand, substantial differences in the formality
of organization existed. Plant engineers and indus-
trial engineers, for example, were rather vague about
their responsibilities, and about the dividing line
between their jobs and other jobs. Similarly, they
reported relatively low reliance on rules and proce-
dures. Production managers, on the other hand,
noted that their jobs were well defined and that rules
and procedures were more important to them.
The effects of these two bases of organization on
differentiation along these three dimensions are
summarized in Exhibit III. Overall, differentiation
was greater between functional specialists at Plant P
than at Plant F.
Integration achieved
While the study found that both plants experienced
some problems in accomplishing integration, these
difficulties were more noticeable at Plant F. Colla-
boration between maintenance and production per-
sonnel and between production and scheduling was a
problem there. In Plant P the only relationship where
integration was unsatisfactory was that between pro-
duction and quality control specialists. Thus Plant P
seemed to be getting slightly better integration in
spite of the greater differentiation among specialists
in that organization. Since differentiation and inte-
gration are basically antagonistic, the only way man-
agers at Plant P could get both was by being effective
at communication and conflict resolution. They
were better at this than were managers at Plant F.
Communication patterns: In Plant P, communica-
tion among employees was more frequent, less formal,
and more often of a face-to-face nature than was the
case with Plant F personnel. One Plant P employee
volunteered:
“Communications are no problem around here.
You can say it. You can get an answer.”
Members of Plant F did not reflect such positive
feelings. They were heard to say:
“Why didn’t they tell me this was going to hap-
pen? Now they’ve shut down the line.”

134 HARVARD BUSINESS REVIEW November–December 1968

This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.

EXHIBIT III. Differentiation in Plants F and P
Dimensions of
differentiation Plant F Plant P
Goal More differentiated Less differentiated
orientation and focused and more diffuse
Time Less differentiated More differentiated
orientation and shorter term and longer term
Formality of Less differentiated, More differentiated,
structure with more formality with less formality
“When we get the information, it is usually too
late to do any real planning. We just do our best.”
The formal boundaries outlining positions that
were more prevalent at Plant F appeared to act as a
damper on communication. The encounters observed
were often a succession of two-man conversations,
even though more than two may have been involved
in a problem. The telephone and written memoranda
were more often employed than at Plant P, where
spontaneous meetings involving several persons
were frequent, usually in the cafeteria.
Dealing with conflict: In both plants, confronta-
tion of conflict was reported to be more typical than
either the use of power to force one’s own position
or an attempt to smooth conflict by “agreeing to dis-
agree.” There was strong evidence, nevertheless,
that in Plant P managers were coming to grips with
conflicts more directly than in Plant F. Managers at
Plant F reported that more conflicts were being
smoothed over. They worried that issues were often
not getting settled. As they put it:
“We have too many nice guys here.”
“If you can’t resolve an issue, you go to the plant
manager. But we don’t like to bother him often with
small matters. We should be able to settle them our-
selves. The trouble is we don’t. So it dies.”
Thus, by ignoring conflict in the hope it would go
away, or by passing it to a higher level, managers at
Plant F often tried to smooth over their differences.
While use of the management hierarchy is one
acceptable way to resolve conflict, so many disagree-
ments at Plant F were pushed upstairs that the hier-
archy became overloaded and could not handle all
the problems facing it. So it responded by dealing
with only the more immediate and pressing ones.
At Plant P the managers uniformly reported that
they resolved conflicts themselves. There was no evi-
dence that conflicts were being avoided or smoothed
over. As one manager said:
“We don’t let problems wait very long. There’s no
sense to it. And besides, we get together frequently
and have plenty of chances to discuss differences
over a cup of coffee.”
HARVARD BUSINESS REVIEW November–December 1968
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
As this remark suggests, the quicker resolution of
conflict was closely related to the open and informal
communication pattern prevailing at Plant P. In
spite of greater differentiation in time orientation
and structure, then, Plant P managers were able to
achieve more satisfactory integration because they
could communicate and resolve conflict effectively.
Performance & attitudes
Before drawing some conclusions from the study
of these two plants, it is important to make two
more relevant comparisons between them—their
effectiveness in terms of the goals set for them and
the attitudes of employees.
Plant performance: As we noted before, the man-
agements of the two plants were aiming at the same
two objectives:
1. Maximizing current output within existing
capabilities.
2. Improving the capabilities of the plant.
Of the two facilities, Plant F met the first objec-
tive more effectively; it was achieving a higher pro-
duction rate with greater efficiency and at less cost
than was Plant P. In terms of the second objective,
however, Plant P was clearly superior to Plant F; the
former’s productivity had increased by 23% from
1963 to 1966 compared with the latter’s increment
of only 3%. One key manager at Plant F commented:
“There has been a three- or four-year effort to im-
prove our capability. Our expectations have simply
not been achieved. The improvement in perfor-
mance is just not there. We are still where we were
three years ago. But our targets for improvements
are realistic.”
By contrast, a key manager at Plant P observed:
“Our crews have held steady, yet our volume is
up. Our quality is consistently better, too.”
Another said:
“We are continuing to look for and find ways to
improve and consolidate jobs.”
Employee attitudes: Here, too, the two organiza-
tions offer a contrast, but the contrast presents a para-
doxical situation. Key personnel at Plant P appeared
to be more deeply involved in their work than did
managers at Plant F, and they admitted more often to
feeling stress and pressure than did their opposite
numbers at Plant F. But Plant F managers expressed
more satisfaction with their work than did those at
Plant P; they liked the company and their jobs more
than did managers at Plant P.
Why Plant P managers felt more involved and had
a higher level of stress, but were less satisfied than
Plant F managers, can best be explained by linking
these findings with the others we have reported.
135
 Study summary
The characteristics of these two organizations are
summarized in Exhibit IV. The nature of the organi-
zation at Plant F seemed to suit its stable but high
rate of efficiency. Its specialists concentrated on their
own goals and performed well, on the whole. The
jobs were well defined and managers worked within
procedures and rules. The managers were concerned
primarily with short-term matters. They were not
particularly effective in communicating with each
other and in resolving conflict. But this was not very
important to achieve steady, good performance, since
the coordination necessary to meet this objective
could be achieved through plans and procedures and
through the manufacturing technology itself.
As long as top management did not exert much
pressure to improve performance dramatically, the
plant’s managerial hierarchy was able to resolve the
few conflicts arising from daily operations. And as
long as the organization avoided extensive problem
solving, a great deal of personal contact was not very
important. It is not surprising therefore that the man-
agers were satisfied and felt relatively little pressure.
They attended strictly to their own duties, remained
uninvolved, and got the job done. For them, this com-
bination was satisfying. And higher management was
pleased with the facility’s production efficiency.
The atmosphere at Plant P, in contrast, was well
suited to the goal of improving plant capabilities,
which it did very well. There was less differentiation
between goals, since the functional specialists to a
degree shared the product goals. Obviously, one dan-
ger in this form of organization is the potential
attraction of specialist managers to total goals to the
extent that they lose sight of their particular goals
and become less effective in their jobs. But this was
not a serious problem at Plant P.
EXHIBIT IV. Observed Characteristics of the Two Organizations
Characteristics Plant F
Differentiation Less differentiation except in
goal orientation
Integration Somewhat less effective
Conflict management Confrontation, but also “smoothing
over” and avoidance; rather restricted
communication pattern
Effectiveness Efficient, stable production; but less
successful in improving plant
capabilities
Employee attitudes Prevalent feeling of satisfaction, but
less feeling of stress and
involvement
136
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
Moreover, there was considerable differentiation
in time orientation and structure; some specialists
worked at the routine and programmed tasks in
operating the plant, while others concentrated on
longer-term problems to improve manufacturing
capability. The latter group was less constrained by
formal procedures and job definitions, and this
atmosphere was conducive to problem solving. The
longer time orientation of some specialists, how-
ever, appeared to divert their attention from main-
taining schedules and productivity. This was a
contributing factor to Plant P’s less effective current
performance.
In spite of the higher degree of differentiation in
these dimensions, Plant P managers were able to
achieve the integration necessary to solve problems
that hindered plant capability. Their shared goals and
a common boss encouraged them to deal directly
with each other and confront their conflicts. Given
this pattern, it is not surprising that they felt very
involved in their jobs. Also they were under stress
because of their great involvement in their jobs. This
stress could lead to dissatisfaction with their situa-
tion. Satisfaction for its own sake, however, may not
be very important; there was no evidence of higher
turnover of managers at Plant P.
Obviously, in comparing the performance of these
two plants operating with similar technologies and in
the same market, we might predict that, because of
its greater ability to improve plant capabilities, Plant
P eventually will reach a performance level at least as
high as Plant F’s. While this might occur in time, it
should not obscure one important point: the func-
tional organization seems to lead to better results in
a situation where stable performance of a routine
task is desired, while the product organization leads
to better results in situations where the task is less
predictable and requires innovative problem solving.
Plant P
Greater differentiation in structure
and time orientation
More effective
Confrontation of conflict; open,
face-to-face communication
Successful in improving plant
capabilities, but less effective in
stable production
Prevalent feeling of stress and
involvement, but less satisfaction
HARVARD BUSINESS REVIEW November–December 1968
 Clues for Managers
How can the manager concerned with the func-
tion versus product decision use these ideas to guide
him in making the appropriate choice? The essential
step is identifying the demands of the task con-
fronting the organization.
Is it a routine, repetitive task? Is it one where inte-
gration can be achieved by plan and conflict managed
through the hierarchy? This was the way the task
was implicitly defined at Plant F. If this is the nature
of the task, or, to put it another way, if management
is satisfied with this definition of the task, then the
functional organization is quite appropriate. While it
allows less differentiation in time orientation and
structure, it does encourage differentiation in goal
orientation. This combination is important for spe-
cialists to work effectively in their jobs.
Perhaps even more important, the functional
structure also seems to permit a degree of integra-
tion sufficient to get the organization’s work done.
Much of this can be accomplished through paper
systems and through the hardware of the production
line itself. Conflict that comes up can more safely be
dealt with through the management hierarchy, since
the difficulties of resolving conflict are less acute.
This is so because the tasks provide less opportunity
for conflict and because the specialists have less dif-
ferentiated viewpoints to overcome. This form of
organization is less psychologically demanding for
the individuals involved.
On the other hand, if the task is of a problem-solv-
ing nature, or if management defines it this way, the
product organization seems to be more appropriate.
This is especially true where there is a need for tight
integration among specialists. As illustrated at Plant
P, the product organization form allows the greater
differentiation in time orientation and structure that
specialists need to attack problems. While encourag-
ing identification with superordinate goals, this orga-
nizational form does allow enough differentiation in
goals for specialists to make their contributions.
Even more important, to identify with product
ends and have a common boss encourages employees
to deal constructively with conflict, communicate
directly and openly with each other, and confront
their differences, so they can collaborate effectively.
Greater stress and less satisfaction for the individual
may be unavoidable, but it is a small price to pay for
the involvement that accompanies it.
The manager’s problem in choosing between prod-
uct and functional forms is complicated by the fact
HARVARD BUSINESS REVIEW November–December 1968
This document is authorized for use only in Prof. Shiva Kakkar's OB-II: Organizational Structure, Design and Change (BOB2BJ22-3), Term - III, BMJ 2022-24 at Xavier Labour Relations
Institute (XLRI) from Jan 2023 to Apr 2023.
that in each organization there are routine tasks and
tasks requiring problem solving, jobs requiring little
interdependence among specialists and jobs requir-
ing a great deal. Faced with these mixtures, many
companies have adopted various compromises
between product and functional bases. They include
(in ascending order of structural complexity):
1. The use of cross-functional teams to facilitate
integration. These teams provide some opportunity
for communication and conflict resolution and also
a degree of the common identification with product
goals that characterizes the product organization. At
the same time, they retain the differentiation pro-
vided by the functional organization.
2. The appointment of full-time integrators or
coordinators around a product. These product man-
agers or project managers encourage the functional
specialists to become committed to product goals and
help resolve conflicts between them. The specialists
still retain their primary identification with their
functions.6
3. The “matrix” or grid organization, which com-
bines the product and functional forms by overlay-
ing them. Some managers wear functional hats and
are involved in the day-to-day, more routine activi-
ties. Naturally, they identify with functional goals.
Others, wearing product or project hats, identify with
total product goals and are more involved in the prob-
lem-solving activity required to cope with long-range
issues and to achieve cross-functional coordination.
These compromises are becoming popular because
they enable companies to deal with multiple tasks
simultaneously. But we do not propose them as a
panacea, because they make sense only for those sit-
uations where the differentiation and integration
required by the sum of all the tasks make a middle
approach necessary. Further, the complexity of inter-
personal plus organizational relationships in these
forms and the ambiguity associated with them make
them difficult to administer effectively and psycho-
logically demanding on the persons involved.
In our view, the only solution to the product versus
function dilemma lies in analysis of the multiple
tasks that must be performed, the differences between
specialists, the integration that must be achieved, and
the mechanisms and behavior required to resolve con-
flict and arrive at these states of differentiation and
integration. This analysis provides the best hope of
making a correct product or function choice or of
arriving at some appropriate compromise solution.
6. See Paul R. Lawrence and Jay W. Lorsch, “New Management Job: The
Integrator,” HBR November–December 1967, p. 142.
137